Stochastic channel model for simulation of mobile ad hoc networks

Abstract

Channel models, applicable to mobile ad hoc network (MANET) simulations, need to be both accurate and computationally efficient. It has been shown that inaccuracies in the channel model can seriously affect various network performance measures. It is essential that the model gives realistic time and spatial variability as the terminals move. Furthermore, the frequency selective effects from multipath propagation must be realistically modelled, so that the effects of different signalling bandwidths are captured correctly. However, to meet the necessary low-complexity constraints current commonly used channel models for network simulations are very simplified. In this study, the authors propose a model structure that is able to capture the essence of the channel characteristics, and to cope with the constraint of low computational complexity. The model describes the channel time and frequency variability between nodes in a MANET. It models the large- and small-scale fading, where the correlation between the fading parameters as well as the spatial correlation is considered. Furthermore, the study presents parameters for the proposed model based on wideband peer-to-peer channel measurements in an urban environment at 300 MHz. When analysing the link and network performance, they show that the proposed channel model describes the channel dynamics appropriately.

Keywords

• mobile ad hoc networks
• fading channels
• communication complexity
• stochastic processes
• stochastic channel model
• mobile ad hoc networks
• MANET
• frequency selective effects
• multipath propagation
• computational complexity
• channel time variability
• frequency variability
• wideband peer-to-peer channel measurements
• channel dynamics
• frequency 300 MHz